Hepatocyte growth factor (HGF), also called scatter factor SF, is a heparin-binding glycoprotein that is secreted as a biologically inert single chain (pro-HGF) and is converted to its bioactive form by targeted protease digestion to an active disulfide-linked heterodimer. HGF is a natural ligand for the c-MET proto-oncogene product of a novel family of heterodimeric receptor tyrosine kinases that include Ron, Sea and Sex. It is a pleiotropic factor derived from the mesenchyme that regulates epithelial, neural, endothelial, muscle and hemopoietic cell growth, motility, morphogenesis and regeneration in many tissues and organs. The importance of HGF is seen in transgenic mice homozygous for a null mutation in the HGF gene. Such mice do not survive beyond fifteen days of embryonic development.
Mature bioactive HFG is a heterodimer consisting of a 60 kD alpha and 30 kD beta chain held together by a single disulfide bond. Structure function analysis indicates that the beta chain of HGF is required for mitogenic activity, whereas the receptor-binding domain is located in the alpha chain. Its primary structure is highly conserved among mouse, rat, human and other species. The alpha chain contains a hairpin loop at its amino terminus and four unique domains known as “kringles”, while its beta chain contains a serine protease-like structure. Hence, HGF is closely homologous to plasminogen, but has no known protease activity due to mutation of the catalytic site.
HGF has been reported to be sequestered in the extracellular matrix (ECM) in vitro as well as in vivo, where it is bound to cell surface heparin sulfate glycosoaminoglycans. In general, HGF mRNA is expressed in stromal cells, whereas HGF receptor expression is mainly detected in epithelial and other parenchymal cells. This pattern suggests that HGF is an important paracrine mediator of the interaction between the parenchymal and stromal components of various tissues both during fetal development and in the maintenance of homeostasis in adult tissues.
Although a great deal is known about the actions of HGF in nonhemopoietic tissues, the role of HGF in the regulation of hematopoiesis, particularly lymphopoiesis, is fragmentary. HGF has been proposed to regulate hematopoiesis in mouse fetal liver and adult bone marrow in vivo, where it apparently can substitute for the stem cell factor (SCF) and c-kit system. HGF is produced by bone marrow (BM) stromal cells and synergizes with IL-3 or GMCSF to support the growth of hemopoietic progenitor cells (HPCs) and myeloid tumor cell lines, all of which express the HGF receptor, c-MET. In the presence of erythropoietin, HGF induces the formation of colonies along the erythroid lineage, whereas in the presence of erythropoietin plus SCF, HGF supports the growth of multipotent colonies. Similarly, upregulation of the HGF receptor on primitive hematopoietic cells may be induced by IL-11; and the synergistic interaction of these two cytokines may result in in vitro colony formation by hemopoietic stem cells (HSCs). However, HGF alone does not appear to stimulate proliferation of hemopoietic precursors. The latter may be attributed to enhancement by HGF of signal transduction by lineage-specific cytokines.
HGF has been found to promote adhesion of HPCs to fibronectin in vitro, and may be involved in a novel paracrine signaling pathway regulating integrin-mediated adhesion and migration of B cells in germinal centers. Messenger-RNA for c-MET has been identified in thymocytes as well as in early B-lineage cells in bone marrow. It is hypothesized that HGF may be involved in a novel paracrine signaling pathway that regulates integrin-mediated adhesion and migration of B-cells in germinal centers. Thus, HGF may be one of the long sought mediators of paracrine interactions between stromal and lymphohematopoietic cells. Furthermore, HGF seems to preferentially affect hematopoietic cells in a window of differentiation between multipotent progenitors and committed precursors. For example, the addition of HGF to fetal thymus organ cultures is known to increase the generation of mature T cells.
Interleukins are a class of proteins that induce growth and differentiation of lymphocytes and hematopoetic stem cells. One interleukin in particular, IL-7, has been demonstrated over the past decade to have an essential role in the development and differentiation of murine pre-B cells.
The nature of IL-7 involvement (if any) at earlier stages of B cell development remains controversial. While it has been proposed that IL-7 is capable of acting on primitive B220− B cell progenitors in the presence of stem cell factor (SCF), most investigators have concluded that the principle B-lineage targets for IL-7 are pro-B cells and pre-B cells. The pre-B cells that do appear in IL-7 KO mice are abnormal as evidenced by their failure to up-regulate or express IL-7R-alpha, TdT and cμ. However, some redundancy may exist between the activities of IL-3, TSLP, and IL-7. Additionally, it has been suggested that the short-term maintenance of pre-pro-B cells, but not pro-B cells, depends on contact-mediated signals from BM stromal cells. Thus, in vivo treatment of mice with anti-IL-7 antibodies eliminates B-lineage subsets as early as the pro-B, but not the pre-pro-B, cell stage; a similar maturational arrest has been observed in mice having disrupted IL-7R-alpha chain genes (IL-7R-alpha −/−); and the Tyr449 to Phe-alpha chain point mutation suggests that the IL-7R transmits distinct signals for cell proliferation and IgH gene rearrangement. In contrast, von Freeden-Jeffry et al. (D. Exp. Med. 181: 1519 (1995)) found that both pre-pro-B cells and pro-B cells are well represented in BM of IL-7 gene-deleted mice; and Pribyl and LeBien (Proc. Nat. Acad. Sci. USA 93: 10348 (1996)) have reported that human B-lineage cells can be generated from fetal precursors in an IL-7-independent manner.
It must be cautioned that the presence of pre-pro-B cells in IL-7R-alpha chain (−/−) mice does not necessarily preclude the involvement of IL-7 at this developmental stage in normal animals. An alternative explanation is that the immediate precursors of pre-pro-B cells do not require an IL-7R-mediated signal to generate pre-pro-B cells. It must also be cautioned that the presence of pro-B cells in IL-7 gene-deleted mice does not exclude a physiological role for IL-7 in early B-lineage development; neither does it preclude the possibility that cytokines other than IL-7 use the IL-7R to stimulate proliferation and differentiation of early B-lineage precursors. Indeed, our recent studies in IL-7 KO mice have demonstrated that IL-7 is essential for upregulation of TdT and IL-7R-alpha chain expression among early pro-B cells and for initiation of cμ expression in late pro-B cells. Therefore, while pro-B cell development occurs in IL-7 KO mice, such development is abnormal. Similar explanations may apply to conflicting reports regarding the need for IL-7 in normal human B cell ontogeny, although important species-specific differences may exist.
In prior studies, the present inventors have demonstrated that serum-free BM stromal cell conditioned medium (CM), as described in Nakumra et al., Nature 342: 440-443 (1989); Rubin et al., Biophysica Acta 1155: 357-371 (1993); and Zarnegar et al., J. Cell Biol. 129: 1177-1180 (1995), selectively stimulates the proliferation of early (TdT−) and late (TdT+) pre-pro-B cells from freshly-harvested rat BM and supports the generation (but not the proliferation) of pro-B cells. Furthermore, adsorption of CM with anti-IL-7 mAb removes this activity, whereas rIL-7 restores this activity to medium conditioned by BM stromal cells from IL-7 gene-deleted mice (−/− CM). Nonetheless, anti-IL-7 mAb is unable to neutralize the pre-pro-B cell growth-stimulating activity in IL-7 (+/+) CM or in rIL-7-supplemented (−/−) CM; and rIL-7, is unable to restore PPBSF activity to IL-7 (+/+) CM that has been adsorbed with anti-IL-7 mAb. The reason for these finding are not explained by the prior art discussed above.